Open-Co-Work, runs locally in an isolated sandbox. How to Install Download and install the latest release for Windows and Mac Open Co-Work Releases OpenSource Repo https://github.com/OpenCoworkAI/open-cowork/releases Once Installed configure the API provider as follows; Select "More Models" Enter your API key from "settings->account" Select "Anthropic"…

Preinstall and configure within WSL "Claude Code". (refer to our Claude Code guide, for instructions on how to Install\Configure Claude Code). Ensure Claude Code is in the path and accessible. Then run the following commands. git clone https://github.com/nexu-io/open-design.git cd open-design corepack enable corepack pnpm --version…

Yes OpenClaw (formerly ClawdBot\Moltbot) is supported with the MiniMax-M2 model. We have a quick start script to get you going. However please exercise caution and follow best practices. VPS, Sandbox, VM and only grant required permissions. Due to the potentially high API costs, this service…

Supported Coding Environments

• Claude Code CLI
• VS Code with Claude Code Extension
• Antigravity with Claude Code Extension
• Gemini CLI
• Codex Desktop (Experimental)
• You may use other tools with the MiniMax series of coding models only.

Installing Claude Code

1. Refer to the Claude Code documentation for installation.

All you need to do is EITHER

Set three environment variables (Linux and WSL), recommend adding to your profile to be persistent.

export ANTHROPIC_BASE_URL="https://chat.ultimateai.org"
export ANTHROPIC_AUTH_TOKEN="<your api key>"
export API_TIMEOUT_MS="3000000"

Or save settings in a config file (recommended for Windows)

1. File located at
   
   • %userprofile%\.claude\settings.json (Windows)
   • ~/.claude/settings.json (Linux)

{
  "env": {
    "ANTHROPIC_AUTH_TOKEN": "<your API key>",
    "ANTHROPIC_BASE_URL": "https://chat.ultimateai.org",
    "API_TIMEOUT_MS": "3000000"
  }
}

Run claude.exe.

Claude Code Extension for VS Code and Antigravity

• Install Claude Code as shown earlier.
• Install Claude Code Extension for VS Code

  

That's it.

Installing Gemini CLI

1. Refer to the Gemini CLI installation, execution, and releases | Gemini CLI Documentation for installation.
2. Set the below environment variables;

GOOGLE_GEMINI_BASE_URL=https://chat.ultimateai.org
GOOGLE_GENAI_USE_VERTEXAI=false
GEMINI_API_KEY=<your api key>

Important when launching gemini append the MiniMax model, example;

gemini –model MiniMax-M2.7

Failure to do this, will result in degraded performance, as Gemini client will try and connect back to Google.  MiniMax usage is not metered and supports all tool use.

Codex Desktop Client

1. Download and Install the Codex Desktop Client
2. Select "Enter API Key"

   

3. Enter your API Key from UltimateAI (settings -> account)

   

4. Add a single line to the top of your config.toml (hint: file-> settings -> open config)

Add this single line to the top of your "config.toml" located in the ".codex" folder.

openai_base_url = "https://chat.ultimateai.org/v1"

Start a NEW chat\window.

Note: For the first message, there may be a few re-tries while the protocol is negotiated.

Obsidian Web Clipper is a convenient way to import chats into Obsidian. {{contentHtml | replace:("/<section><h2>.*?<\/h2>/g":"<section>", "<p><span>":"★BOT said<p><span>", "/<p>\S*</p>\s*<pre/g":"<pre") | markdown | blockquote | replace:("> ★BOT said\n> \n> ":"\n> [!bot]- ", "> ★User:★\n>":"\n> [!user]- Me")}} Optional (Set a Template Trigger), eg: https://.obsidianaitools.com, https://*.ultimateai.org, ect.. for the…